»ö« | perl6-projects.org/ | evalbot: 'perl6: say 3;' | irclog: irc.pugscode.org/ | UTF-8 is our friend!
Set by moritz_ on 17 June 2009.
00:12 edenc joined
edenc p6eval: say 3 00:13
perl6: say 3 00:14
p6eval pugs, rakudo 6c43f9: OUTPUT«3␤»
..elf 27271: ( no output )
edenc p6eval: my @a = 1,2...[+]; say @a[5]; 00:15
perl6: my @a = 1,2...[+]; say @a[5]; 00:19
p6eval elf 27271: OUTPUT«Parse error in: /tmp/3FI23Tld4g␤panic at line 1 column 0 (pos 0): Can't understand next input--giving up␤WHERE: my @a = 1,2...[+]; say @a[5];␤WHERE:/\<-- HERE␤ STD_red/prelude.rb:99:in `panic'␤ STD_red/std.rb:76:in `scan_unitstopper'␤ STD_red/std.rb:224:in `comp_unit'␤
..STD_red/…
..rakudo 6c43f9: OUTPUT«Statement not terminated properly at line 2, near "...[+]; sa"␤in Main (src/gen_setting.pm:3225)␤»
..pugs: OUTPUT«*** ␤ Unexpected "];"␤ at /tmp/OaGIz91dsB line 1, column 17␤»
00:19 cognominal left
TimToady rakudo: 1,2,...&[+] 00:24
p6eval rakudo 6c43f9: OUTPUT«Statement not terminated properly at line 2, near "&[+]"␤in Main (src/gen_setting.pm:3225)␤»
TimToady rakudo: 1,2,...&infix:<+>
p6eval rakudo 6c43f9: OUTPUT«Statement not terminated properly at line 2, near ":<+>"␤in Main (src/gen_setting.pm:3225)␤» 00:25
TimToady rakudo: 1,2...&infix:<+>
p6eval rakudo 6c43f9: ( no output )
TimToady rakudo: say 1,2...&infix:<+>
p6eval rakudo 6c43f9: ( no output )
TimToady rakudo: say 1,2...{ (state $i)++ < 100 ?? $^a + $^b !! () } 00:26
p6eval rakudo 6c43f9: OUTPUT«Use of uninitialized
..value␤123581321345589144233377610987159725844181676510946177112865746368750251213931964183178115142298320401346269217830935245785702887922746514930352241578173908816963245986102334155165580141267914296433494437701408733113490317018363119032971215073480752697677…
00:29 athomason left 00:39 dduncan joined, dduncan left 00:44 snarkyboojum joined 00:45 xinming_ joined 00:46 synth left 00:51 xinming left, synth joined 00:57 xinming_ left 00:58 cognominal joined, xinming joined 00:59 synth left 01:00 justatheory left 01:18 DemoFreak left 01:26 last joined 01:27 SmokeMachine joined 01:28 drbean_ joined 01:32 twigil joined 01:34 SmokeMachine left 01:35 SmokeMachine joined, payload left 01:36 SmokeMachine left, SmokeMachine joined, wknight8111 left 01:37 SmokeMachine left, SmokeMachine joined 01:38 SmokeMachine left 01:39 lichtkind_ left 01:40 drbean left 01:46 mkfort left 01:48 payload joined 01:56 agentzh left 01:58 molaf_xx left 01:59 molaf_xx joined 02:01 snarkyboojum left
amoc rakudo: my $c = -> $x { state $i++; say "$x: $i" }; $c('a'); $c('b'); 02:13
p6eval rakudo 6c43f9: OUTPUT«a: 1␤b: 2␤»
02:21 icwiener left 02:32 snarkyboojum joined 02:50 nihiliad joined 02:52 snarkyboojum_ joined 02:57 snarkyboojum left 02:59 Limbic_Region left 03:11 zamolxes joined 03:14 viklund left 03:15 FurnaceBoy left 03:16 agentzh joined 03:29 azawawi joined
azawawi good morning 03:29
03:32 last left, cls_bsd left 03:35 cls_bsd joined 03:44 jaldhar joined
skids do we have a working pastebot again? 03:47
03:48 mkfort joined
skids nopaste? 03:48
TimToady lisppaste3? 03:53
lisppaste3: uri? 03:54
eternaleye paste?
Eh, I'll fetch from backlog 03:55
TimToady irclog.perlgeek.de/perl6/2009-06-26#i_1268441
eternaleye paste.lisp.org/new/perl6 03:57
03:58 KyleHa joined
eternaleye Lesson learned: psql and grep are slower than infrared clogs 03:58
skids Ah, well, already have in in gist now at this point, but thanks. 04:02
Can anyone topic that? 04:03
eternaleye TimToady: Aha, lisppaste3 responds to 'url?' rather than 'uri?'
lisppaste3: url?
lisppaste3 To use the lisppaste bot, visit paste.lisp.org/new/perl6 and enter your paste.
skids phenny: tell ruoso gist.github.com/136886 04:09
phenny skids: I'll pass that on when ruoso is around.
04:14 synth joined
pugs_svn r27272 | lwall++ | [Cursor.pmc] fake up a YOU_ARE_HERE for .pm files to dump .yml 04:28
r27272 | lwall++ | [Cursor.pmc] don't add & names for enums or exported constants
r27272 | lwall++ | [Cursor.pmc] .yml doesn't include its setting, and hides GLOBAL as UNIT::GLOBALish
r27272 | lwall++ | [DEBUG.pm] typo
r27272 | lwall++ | [CORE.setting] change some classes to roles and subsets now that those work
r27272 | lwall++ | [CORE.setting] fake up DEBUG module as enum for now
r27272 | lwall++ | [STD.pm] fix SETTING and UNIT semantics
r27272 | lwall++ | [STD.pm] don't add & names in constant context
r27272 | lwall++ | [STD.pm] reconstruct CORE and GLOBAL correctly from SETTING that maybe isn't CORE
r27272 | lwall++ | [STD.pm] don't localize $*CORE, $*GLOBAL, and $*UNIT since they're used outside .parsefile
r27272 | lwall++ | [STD.pm] don't create new pad for role's signature before you know there is one
r27272 | lwall++ | [Makefile] don't fail snapshot on unimportant transients
azawawi wow 04:29
overwhelming commit
TimToady have the appropriate amount of fun... 04:30
04:30 KyleHa left, KyleHa joined
TimToady std: 72 04:30
p6eval std 27271: OUTPUT«ok 00:02 35m␤» 04:31
TimToady I didn't quite get to the point of teaching use how to pull the .pm's .yml back in, however 04:32
KyleHa perl6: my $x = 42;my $y = \$x; say $y.WHAT.perl
p6eval elf 27272: OUTPUT«Undefined subroutine &GLOBAL::prefix__92 called at (eval 125) line 4.␤ at ./elf_h line 5881␤»
..pugs: ( no output )
..rakudo 6c43f9: OUTPUT«Int␤»
TimToady that's for tomorrow's airplane ride, I guess, and then I can kick Test and DEBUG out of CORE.setting 04:33
azawawi meanwhile i will get viv syntax highlighting and rename variable refactoring up and running :) 04:34
i really like viv
TimToady it has it's ups and downs
KyleHa rakudo: my $x = 42;say $x.WHAT.perl;
p6eval rakudo 6c43f9: OUTPUT«Int␤»
KyleHa An Int and a reference to an Int are both Int?
azawawi TimToady: what are viv 's problems? 04:35
TimToady it's not done yet :P
KyleHa: a capture containing a single item will behave as that single item in scalar context 04:36
azawawi no problem... i'll take it as it is if it works :)
TimToady std: 72 04:37
p6eval std 27271: OUTPUT«ok 00:02 35m␤»
KyleHa Are captures described in S02?
azawawi breakfast & 04:38
TimToady they're kinda described all over the place
I recommend a recursive grep of the Spec dir 04:39
KyleHa Yeah, I did that...that's why I wasn't sure what the "one place" was.
Turns out there isn't a one place.
TimToady well, I'm only going to get 4 hrs of sleep now, so I'd better try to zzz off 04:41
04:42 KyleHa1 joined
TimToady stupid how you have to get up early to fly west to a later time zone... 04:42
04:42 KyleHa left
TimToady hits the hay & 04:42
std: 72
p6eval std 27272: OUTPUT«ok 00:02 35m␤»
KyleHa1 Hotel Wi-Fi is not exactly solid.
TimToady std: Num::pi 04:43
p6eval std 27272: OUTPUT«Undeclared name:␤ Num::pi used at 1 ␤ok 00:02 35m␤»
TimToady std: pi
p6eval std 27272: OUTPUT«ok 00:02 35m␤»
TimToady strange that it exports pi and not Num::pi
oh well, something else for tomorrow 04:44
zzz & # thunk
04:53 Molaf joined 04:57 KyleHa joined, KyleHa1 left 04:58 KyleHa left, KyleHa1 joined
eternaleye re pastebots, I just sent gist.github.com/136901 on GitHub's contact form, so maybe someday we'll have a gist-lister bot 04:58
04:59 KyleHa1 left 05:02 s1n left, s1n joined 05:04 s1n left, s1n joined
amoc std: [+] 05:07
p6eval std 27272: OUTPUT«ok 00:02 36m␤»
05:15 azawawi left 05:17 azawawi joined
azawawi @tell literal i am having grok 0.06 win32 installation issues, please see gist.github.com/136907 05:21
lambdabot Consider it noted.
05:27 azawawi left 05:29 Trey left 05:38 justatheory joined 05:40 meppl joined 05:47 justatheory left 05:49 M_o_C joined 05:51 ejs joined 05:54 skids left
pugs_svn r27273 | kyle++ | [t/spec] Added autoincrement-range.t with a lot of overlap with autoincrement.t 05:54
05:54 KyleHa joined 06:05 KyleHa left, KyleHa1 joined 06:06 KyleHa1 left 06:07 mberends joined 06:09 [bjoern] left 06:15 ejs left 06:28 KyleHa joined
pugs_svn r27274 | kyle++ | [t/spec] Bring autoincrement.t in line with autoincrement-range.t 06:28
KyleHa Since I broke out autoincrement-range.t, I'm thinking that needs to be added to rakudo's t/spectest.data somewhere. 06:29
06:38 ejs0 joined
pugs_svn r27275 | kyle++ | [t] Fix lies in the README file. 06:40
06:41 Avada joined
KyleHa If that's not enough damage for the weekend, maybe I can pick it up again Sunday night when I'll be even more tired. Nyuck nyuck nyuck. 06:41
06:42 KyleHa left 06:48 finanalyst joined
amoc rakudo: class A { method new () { say "nth" }; method new ($x) { $x.say } }; my $o = A.new; 06:52
p6eval rakudo 6c43f9: OUTPUT«too few arguments passed (1) - 3 params expected␤in method A::new (/tmp/NpWfihS7A0:2)␤called from Main (/tmp/NpWfihS7A0:2)␤»
amoc reading syns if he is wrong.
rakudo: class A { method new #{no-()}{ say "nth" }; method new ($x) { $x.say } }; my $o = A.new; 06:54
p6eval rakudo 6c43f9: OUTPUT«too few arguments passed (1) - 3 params expected␤in method A::new (/tmp/Bk6hP9Eqsa:2)␤called from Main (/tmp/Bk6hP9Eqsa:2)␤»
06:59 nihiliad left 07:02 xinming_ joined 07:03 nihiliad joined
amoc rakudo: class A { submethod BUILD() { say "nth" }; submethod BUILD($x) { $x.say } }; my $o = A.new; 07:03
p6eval rakudo 6c43f9: OUTPUT«too few arguments passed (1) - 3 params expected␤in submethod A::BUILD (/tmp/ZlHwBx56ry:2)␤called from Main (/tmp/ZlHwBx56ry:2)␤»
amoc rakudo: class A { submethod BUILD($x) { $x.say }; }; my $o = A.new('hi'); 07:04
p6eval rakudo 6c43f9: OUTPUT«too few arguments passed (1) - 3 params expected␤in submethod A::BUILD (/tmp/TaM1q4vlXJ:2)␤called from Main (/tmp/TaM1q4vlXJ:2)␤»
07:05 nihiliad left
amoc std: class A { submethod BUILD($x) { $x.say }; }; my $o = A.new('hi'); # it may be at least not a grammatical err? 07:08
p6eval std 27275: OUTPUT«ok 00:03 38m␤»
07:15 xinming left 07:21 Avada left
amoc rakudo: class A { has $.bar is rw; submethod BUILD($.bar = 5) {} }; my $o = A.new(bar=>1); $o.bar.say 07:25
p6eval rakudo 6c43f9: OUTPUT«Use of uninitialized value␤␤»
amoc does not report rakudobug (as there is "rakudo todo" flag in t/spec/S12-attributes/instance.t) 07:26
moritz_ »ö« | perl6-projects.org/ | nopaste: paste.lisp.org/new/perl6 | evalbot: perl6: say 3;' | irclog: irc.pugscode.org/ | UTF-8 is our friend! 07:31
07:38 DemoFreak joined 07:39 zamolxes left
moritz_ rakudo: PAST::Val.new() 07:39
p6eval rakudo 6c43f9: ( no output )
07:57 Ryan52 joined 08:00 iblechbot joined
amoc iblechbot: help 08:04
iblechbot: ping
08:14 ejs joined 08:15 ejs left 08:16 Chillance joined 08:17 amoc left 08:19 ejs0 left 08:23 viklund joined 08:28 ejs joined 08:34 amoc joined 08:35 Su-Shee joined
Su-Shee good morning. 08:35
08:35 snarkyboojum_ left 08:41 jlmoko left 08:56 finanalyst left, PZt joined 08:59 ejs1 joined 09:00 ejs left 09:02 ejs1 left 09:11 [particle]1 joined 09:13 beggars joined 09:24 ejs0 joined 09:27 [particle] left 09:29 jlaire joined 09:34 icwiener joined
amoc rakudo: class A { method foo() {say 'no argument~'}; method foo($a) { $a.say }}; my $o = A.new; $o.foo(1); $o.foo(); 09:38
p6eval rakudo 6c43f9: OUTPUT«1␤too few arguments passed (1) - 3 params expected␤in method A::foo (/tmp/llOR59FON7:2)␤called from Main (/tmp/llOR59FON7:2)␤»
amoc std: class A { method foo() {say 'no argument~'}; method foo($a) { $a.say }}; my $o = A.new; $o.foo(1); $o.foo();
p6eval std 27275: OUTPUT«ok 00:03 38m␤»
amoc pugs: class A { method foo() {say 'no argument~'}; method foo($a) { $a.say }}; my $o = A.new; $o.foo(1); $o.foo(); 09:39
p6eval pugs: OUTPUT«1␤*** No such method in class A: "&foo"␤ at /tmp/dsSNGtRI1l line 1, column 100-108␤»
amoc er... 09:40
pugs: class A { method foo() {say 'no argument~'}; method foo($a) { $a.say }}; my $o = A.new; $o.foo('one argument');
p6eval pugs: OUTPUT«one argument␤»
amoc pugs: class A { method foo() {say 'no argument~'}; method foo($a) { $a.say }}; my $o = A.new; $o.foo(); #no argument
p6eval pugs: OUTPUT«*** No such method in class A: "&foo"␤ at /tmp/kSoItAAMb5 line 1, column 89-97␤» 09:41
amoc pugs: class A { method foo() {say 'no argument~'} }; my $o = A.new; $o.foo(); #no argument + one dispatch 09:42
p6eval pugs: OUTPUT«no argument~␤»
amoc check test-suites and tickets
moritz_ amoc: the error is that rakudo should complain about re-defintion of non-multi method 09:51
09:57 pmurias joined
pmurias skids_: ping 09:57
10:00 payload left 10:23 Muixirt joined 10:34 ejs0 left
amoc is back 10:49
moritz_ is gone for the weekend 10:50
amoc moritz_: have a good weekend! 10:52
amoc checks to submit
TimToady has just settled in at gate A3 10:57
amoc submitted [perl #67024] 11:03
11:16 iblechbot left
TimToady hmm, I'm thinking constant wants to be "our" rather than "my", actually 11:26
otherwise, after class Num { constant pi is export = 3; }, you can get to pi but can't get to Num::pi 11:27
pmurias perl6: sort(1,2,3,4) 11:39
p6eval elf 27275: OUTPUT«Undefined subroutine &GLOBAL::sort called at (eval 122) line 3.␤ at ./elf_h line 5881␤»
..pugs, rakudo 6c43f9: ( no output )
pmurias perl6: say sort(1,2,3,4)
p6eval elf 27275: OUTPUT«Undefined subroutine &GLOBAL::sort called at (eval 123) line 3.␤ at ./elf_h line 5881␤»
..pugs, rakudo 6c43f9: OUTPUT«1234␤»
pmurias perl6: say sort([3,2,1,5]) 11:40
p6eval pugs: OUTPUT«1235␤»
..elf 27275: OUTPUT«Undefined subroutine &GLOBAL::sort called at (eval 123) line 3.␤ at ./elf_h line 5881␤»
..rakudo 6c43f9: OUTPUT«3 2 1 5␤»
pmurias perl6: say sort({^a cmp $^b},[3,2,1,5])
p6eval elf 27275: OUTPUT«Undefined subroutine &GLOBAL::cmp called at (eval 128) line 3.␤ at ./elf_h line 5881␤»
..rakudo 6c43f9: OUTPUT«3 2 1 5␤»
..pugs: OUTPUT«3 2 1 5<SubBlock(<anon>)>␤»
11:42 fridim joined
TimToady std: Num::pi 11:43
p6eval std 27275: OUTPUT«Undeclared name:␤ Num::pi used at 1 ␤ok 00:02 35m␤»
TimToady std: pi
p6eval std 27275: OUTPUT«ok 00:02 35m␤»
pugs_svn r27276 | lwall++ | [STD] make 'constant' have 'our' semantics 11:44
TimToady std: Num::pi 11:47
p6eval std 27275: OUTPUT«Undeclared name:␤ Num::pi used at 1 ␤ok 00:02 35m␤»
TimToady std: Num::pi 11:52
p6eval std 27275: OUTPUT«Undeclared name:␤ Num::pi used at 1 ␤ok 00:02 35m␤»
11:56 vv31 joined
TimToady std: Num::pi 11:57
p6eval std 27276: OUTPUT«Undeclared name:␤ Num::pi used at 1 ␤ok 00:02 35m␤»
TimToady hmm, need to force recompile of setting before that'll work 11:58
well, they're boarding, so I'll have to shut down soon
pugs_svn r27277 | lwall++ | [CORE.setting] just for a recompile 11:59
TimToady bbl &
12:00 ruoso joined
ruoso Hello! 12:07
phenny ruoso: 04:09Z <skids> tell ruoso gist.github.com/136886
amoc rakudo: my @list = 1..4; say 2 ~~ @list; 12:08
ruoso skids_, hi
p6eval rakudo 6c43f9: OUTPUT«0␤»
ruoso @fisl10... will give a Perl 6 talk in about 3 hours 12:09
12:13 xomas left
Su-Shee ruoso: any live stream? :) 12:13
ruoso Su-Shee, I think there is... but the talk is in portuguese
12:14 xomas joined
Su-Shee ruoso: which I'm sure will sound great, because portuguese has a nice ring.. but ok. I won't understand. :) 12:15
12:16 skids joined
mberends wishes ruoso a good talk, and looks forward to learning some Portuguese at YAPC::EU 12:17
12:17 wknight8111 joined
Muixirt is there some workaround for the lack of 'buf' or 'buf8' in rakudo? 12:19
ruoso skids, hi 12:21
skids ruoso: hey
Muixirt: I think Buf may be higher on pmichaud's TODO list nowadays. At least he spent some time getting the specs straight with TimToady a week or so ago. 12:22
ruoso: so did I get most of that right?
Muixirt skids, thanks 12:23
skids, can stuff my binary data into strings without the unicode ballyhoo that mangles it? 12:25
amoc it seems there is no ( Num ~~ Array ) case in smart-match operator in spec.. '~ ' 12:26
ruoso: wishes good talk ! 12:27
12:31 payload joined
skids Muixirt: there's really no good way to handle packed binary data yet. 12:31
Except maybe if you can get it into an array as nums.
(believe me, I've tried :-) 12:33
ruoso skids, yes... it is... you could even commit it already
amoc, thansk
skids, the writing is correct... the struct SMOP__Object simply dfines the header for every SMOP__Object-compatible structure 12:34
skids, a weak reference is an object that doesn't count in the reference counting 12:36
skids, to solve circular-reference issues
skids, and yes, the capture owns the object that you put inside it...
skids OK, so the object may disappear with the reference intact.
(if you are not careful)
ruoso skids, and when that happens, your weakref starts pointing to false 12:37
skids Ah, so a REFERENCE(weakref) would work, too, it's just a proxy object that the GC will route to false later. 12:38
So how does one get ownership back from a returned capture? 12:39
12:39 amoc__ joined 12:40 amoc__ is now known as am0c
skids Or I guess I mean, how does one extract a non-capture object from a capture... that's probably something I will find with more reading. 12:40
How about #4?
ruoso skids, to get ownership you just need to REFERENCE, then you also has the ownership 12:42
skids, initially you would need to send MESSAGE to it, 12:43
but that obviously creates a circularity problem
so you can check if the capture you received is a native capture
and, in that case, use the C functions to manipulate it
and you can know if it is a native capture by looking at the responder interface
pmurias ruoso: hi 12:46
ruoso hi pmurias
skids, in #4, the capture owns the object inside it, yes... 12:47
pmurias skids: re #3 most of the time MESSAGE should returns a native false, and the actual return value is passed by setr 12:51
skids What I mean by #4 is, how many times would the code that got the capture release, and how many times would the code that sent the message release?
pmurias 1-1 12:52
skids OK, so it is a counting not total transfer of all ownership. Just making sure.
I'll have to read further to get to setr, but I'll put something vaguely correct in there until I do. 12:54
12:56 amoc left, payload left
skids But first, I have to clean up after my cats :-P 12:56
pmurias ruoso: have you seen the new hack used to export stuff out of Test.pm? 13:00
ruoso not yet... getting everything set up to the talk I'll give...
pmurias ruoso: what's your talk title? 13:02
ruoso Perl 6 - em busca do santo graal, the quest for the hole grail, in english 13:03
13:06 fridim left 13:12 wknight8111 left, last joined 13:13 wknight8111 joined
skids Does it turn out that the grail is actually a woman (or, female butterfly?) 13:19
13:33 Trey joined 13:35 mberends left 13:37 finanalyst joined 13:39 snarkyboojum joined 13:54 gabiruh joined
skids Hrm, captures may survive a MESSAGE, correct? So ownership isn't really transferred it is copied? 13:57
13:58 M_o_C left 14:00 M_o_C joined 14:01 a3r0 joined 14:05 Lorn left
jnthn hi all - back from vacation 14:09
pmurias skids: 1 unit of ownership is transfered 14:15
14:16 M_o_C left
pmurias skids: the reference counter represent the number of owners an object has 14:16
pugs_svn r27278 | wayland++ | While I'm usually happy to let Larry be the benevolent dictator, this time I must put my 14:21
r27278 | wayland++ | foot down. The repetition of my name (implying that I'm twice as important as the other
r27278 | wayland++ | person mentioned), while flattering, and while complying with the spirit of his dictum of
r27278 | wayland++ | treating everyone equally (in that it treats those with more merit ;) more highly than
r27278 | wayland++ | others), does not comply with the letter of the aforesaid dictum, in that my name appears
r27278 | wayland++ | twice, and the name of the mere benevolent dictator for life appears only once. In order
r27278 | wayland++ | that the situation might be remedied immediately, I hereby make this commit to our
r27278 | wayland++ | repository.
skids pmurias: got it. Just to check, though, captures don't get consumed by MESSAGE, right, unless the code calling the MESSAGE does not refer to them beyond that point? 14:23
(that is, the code still has to release the capture itself) 14:24
ruoso skids, usually the code that receives a capture should release it at the end 14:34
skids Ah, that's a good thing to know. 14:35
ruoso jnthn, welcome back
in a few minutes we'll be live from the Perl 6 talk I'll be giving 14:36
Trey ruoso: where?
skids If it does not release the capture for some reason, it should not release the objects that were inside, or should it?
ruoso Porto Alegre, Brasil
Inernational Free Software Forum - FISL 14:37
7k atendents this year
skids (and, corollary to that, it need only release the capture, not the individual objects?)
Trey nice. slides online?
in English, then, I was going to say something em Português :) 14:38
ruoso Trey, the slides are not yet online... they are currently in portuguese, but I now have a policy of bilingual posting, so I should translate the slides as well soon 14:39
14:40 mjk joined
ruoso skids, if it gets one of the objects inside the capture, it will own that object, so it should release it as well... 14:40
skids so during a call both the capture object and the code own each object? 14:42
And most cases, the caller will assume that the callee will release the capture object?
Trey rakudo: sub baz { say "$^a!" }; try { baz 'one', 'two'; CATCH { default { say 'whoops' } }} 14:43
p6eval rakudo 6c43f9: OUTPUT«whoops␤»
Trey rakudo: sub baz { say "$^a!" }; { baz 'one', 'two'; CATCH { default { say 'whoops' } }}
p6eval rakudo 6c43f9: OUTPUT«whoops␤too many arguments passed (2) - 1 params expected␤in sub baz (/tmp/kbM6jjGuyy:1)␤called from Main (/tmp/kbM6jjGuyy:2)␤»
skids Or, it has to explicitly get the object out of the capture... and until then it does not own the object.
Trey Is that correct? I thought the presence or absence of "try" was not supposed to have any effect when a CATCH block was present
xinming_ might be a dumb question, How complete is rakudo perl 6 for now please? I mean, the completiness. 14:44
Trey xinming_: rakudo.org/status
xinming_ Trey: thanks 14:46
It seems, It catches up with old pugs implementation now.
14:46 fridim joined
skids xinming_: in most things. pugs still has laziness, rakudo not yet. 14:50
14:51 Limbic_Region joined, PacoLinux left
xinming_ skids: Because It's based on haskell. :-) 14:51
14:54 pmichaud_ is now known as pmichaud
skids skids: yep, pugs got laziness for free. rakudo needs to work for it. 14:54
Trey pmichaud: see last two rakudo's ^^ is my understanding of C<try> wrong?
pmichaud Trey: I suspect we don't automatically convert CATCH blocks into try's 14:55
14:56 mjk left, |mjk| joined
Trey once i figure out how the tests are laid out now, i'll add a test, my commit bit still works, but it's been three years since I've touched the test suite :) 14:57
jnthn pmichaud: o/
14:58 FurnaceBoy joined
pmichaud hello, jnthn! 14:58
back from travels, or still on them? 14:59
jnthn Back.
pmichaud excellent. I'll be back on Tue.
jnthn OK. How's things? I've been almost entirely offline...
pmichaud (I'll be around off-and-on between now and then, but currently I'm on "vacation")
things are fairly well. Lots of little role and class issues have been making it to RT -- I'm hoping you can look at some of those. 15:00
jnthn *nod*
pmichaud I've been unable to switch PMCs to use 'rw' instead of 'readonly' -- Parrot segfaults and/or has memory corruption issues.
jnthn Ouch.
gc-related or not or unknown? 15:01
pmichaud even simply adding the property (but not using it, or getting rid of the 'readonly' property) is sufficient to mess things up.
gc-related; or at least the -G flag causes the problems to disappear.
jnthn OK, not good. 15:02
pmichaud I started closing tickets on Thursday, but getting operator definition in setting ended up taking a long time (due to spectest slowness) 15:03
anyway, I think we can start defining operators in setting
(there are already a few there)
jnthn OK, sounds good. I will git pull, plus be reading the spec commits over the next couple of days. 15:04
pmichaud okay, good plan.
jnthn Try and get caught up with what's been happening.
pmurias skids: the calle has to explicitly get out the object from the capture 15:05
pmichaud Not a whole lot of significance changed while you were gone -- I kept running into Parrot GC issues.
and preparation for pvmw and yapc::na
jnthn OK, did they happen yet? 15:06
pmichaud yes
jnthn OK. How did they go?
pmichaud pvmw was last weekend, yapc::na was mon-wed, hackathon on thurs (when I worked on operators in setting)
they all went really well. We had very good participation at pvmw
I need to write up my summary, although you can probably find whiteknight's summary on the web
jnthn ok, great 15:07
15:07 snarkyboojum left 15:08 PacoLinux joined
skids pmurias: ok, so the only ownership that is transferred is that of the aggregate capture object. Much clearer now. 15:09
ruoso Hey guys!! you're live now at FISL
say hello to Porto Alegre
pmichaud hello, Porto Alegre!
skids hola! Or is that only Spanish?
15:10 beggars left
ruoso hola is spanish... you can say "OI" 15:10
skids "OI".say 15:12
Aren't we a talkative bunch now? :-) 15:13
15:14 |mjk| left 15:15 gfldex is now known as Guest93618
pmurias skids: it might be helpfull to look at the code a bit too 15:17
skids Yeah, this last section really requires more reading than what I've done so far. 15:18
But I think I can get it mostly right and then go back and fix details.
However, it is good to try at least before reading the code, that way, I get the overview documentation to answer questions people who have not yet read the code might have. 15:21
(which I might overlook otherwise)
15:23 [bjoern] joined 15:24 Guest93618 left, azawawi joined
azawawi hi 15:24
xinming_ BTW, it seems, feather is running out of swap, anyone here have priviledge to enlarge the swap? 15:27
15:28 PacoLinux left 15:32 PacoLinux joined
pugs_svn r27279 | skids++ | Grammar, many clarifications, more explicit. 15:33
r27279 | skids++ | Will revisit after reading more code.
15:38 PacoLinux left 15:39 PacoLinux joined
pmurias azawawi: hi 15:39
azawawi pmurias: hey! I see mildew is moving nicely :) 15:41
15:48 PacoLinux left
[bjoern] rakudo: say (::&say) 15:51
p6eval rakudo 6c43f9: OUTPUT«ResizablePMCArray: Can't pop from an empty array!␤in Main (src/gen_setting.pm:2183)␤»
15:53 PacoLinux joined
[bjoern] rakudo: ... 15:53
p6eval rakudo 6c43f9: OUTPUT«Can't return outside a routine␤in Main (/tmp/nIP8tfmBs4:2)␤»
[bjoern] rakudo: say nothing 15:55
p6eval rakudo 6c43f9: OUTPUT«Null PMC access in isa()␤in Main (/tmp/lrWluvNZM0:2)␤»
15:57 brunov joined, lichtkind joined, azawawi left
pugs_svn r27280 | azawawi++ | [S:H:P6] experiment with viv's based coloring 16:00
16:04 REPLeffect left 16:07 a3r0 left
pugs_svn r27281 | azawawi++ | [S:H:P6] Move find variable declaration and rename variable into their own functions... 16:08
16:14 iblechbot joined, PacoLinux left 16:16 PacoLinux joined 16:20 PacoLinux left 16:22 PacoLinux joined 16:26 icwiener left 16:28 jonathanturner joined 16:29 ruoso left 16:35 maettu joined 16:41 lichtkind left 16:43 Psyche^ joined, icwiener joined
pugs_svn r27282 | azawawi++ | [S:H:P6] A bit of ansi coloring fun. 16:44
pmurias skids: btw if you have any questions about the code feel free to ask 16:47
pugs_svn r27283 | azawawi++ | [S:H:P6] add use lib 'lib' 16:48
r27284 | azawawi++ | [S:H:P6] Fixed ansi coloring to actually work. 16:50
16:52 icwiener left 16:54 ChanServ sets mode: +o TimToady 16:55 icwiener joined 16:56 ChanServ sets mode: -o TimToady 16:57 finanalyst left 16:59 Patterner left, Psyche^ is now known as Patterner, Chillance left 17:00 _jaldhar joined 17:04 icwiener left 17:05 icwiener joined 17:12 fronty joined, gbacon joined 17:15 justatheory joined 17:20 synth left
pugs_svn r27285 | pmurias++ | [re-mildew] 17:22
r27285 | pmurias++ | added infix >
r27285 | pmurias++ | added example/qsort.pl
r27286 | pmurias++ | [re-mildew] optional positionals $foo? are supported
17:24 jaldhar left
pugs_svn r27287 | azawawi++ | [S:H:P6] more coloring 17:24
17:25 kane__ joined 17:39 mberends joined 17:43 Limbic_Region left 17:47 justatheory left 17:48 brunov_ joined 17:52 brunov_ left 18:03 Molaf left, Molaf joined 18:05 loget joined
loget evalbot: perl6: eval('say 3;'); 18:06
evalbot: perl6: say 3
evalbot: perl6: 'say 3;'
evalbot: perl6: say 3;
doesn't work
huf rakudo: say 3; 18:07
p6eval rakudo 6c43f9: OUTPUT«3␤»
loget oh
huf std: say 3;
p6eval std 27287: OUTPUT«ok 00:02 36m␤»
huf ;)
loget rakudo: eval('say 3;'); 18:08
p6eval rakudo 6c43f9: OUTPUT«3␤»
loget rakudo: eval('label: say 3;');goto label;
p6eval rakudo 6c43f9: OUTPUT«Could not find non-existent sub label␤»
loget -_-
I think that's a bug
rakudo: eval('label: say 3;goto label;'); 18:09
p6eval rakudo 6c43f9: ( no output )
arnsholt Is schroeer on github someone in here? 18:11
18:14 jlaire_ joined 18:16 loget left 18:18 jlaire left 18:19 jlaire_ is now known as jlaire 18:20 bpa joined 18:26 masak joined
masak ahoj, butterflies! 18:26
today's question: (a repeat from yesterday, I think. I missed the answer) -- what happens when a protoregex are defined in a grammar A, grammar B derives from grammar A, and a protoregex with the same name is defined in grammar B? 18:32
will there be one or two protoregexes in grammar B? 18:33
and -- regardless of the answer to that question -- how do I get the other behaviour?
18:41 nihiliad joined 18:42 fronty left 18:52 ejs0 joined 18:54 meppl left 18:59 M_o_C joined 19:00 m-i-l-a-2 is now known as m-i-l-a-n, twigil left 19:02 payload joined
jnthn masak: my first guess would be that you might get inheritance semantics along the lines of multi methods. 19:04
Though that's a guess rather than a "I can remember a spec reference"
masak jnthn! \o/ you're back!
[bjoern] Is there a declarative way to match on the difference between two regexps? As in, "regex Variable { <Name> but not <ReservedName> }"
jnthn lookahead maybe 19:05
regex Variable { <!before <ReserveName> <ws>> <Name> } 19:06
masak [bjoern]: <!before <ReservedName>>
jnthn masak: Yes, back today
masak jnthn: how was vacation?
jnthn masak: Great, though very hot at times. 19:07
Especially the day I was treking up a sulphur-spewing volcano. 19:08
[bjoern] <Name> !~~ ^ <ReservedName> $ ?
19:08 dukeleto_ joined
jnthn masak: How's things been for you? 19:09
masak jnthn: oh, as usual. a bit hectic this week, but otherwise fine.
I've missed the jnthn-related activity on #perl6 and use.perl.org, so it's nice to have you back. :) 19:10
[bjoern] How about complement ala "Anything but a <Name>" (other than writing it in the same way as before) 19:11
masak jnthn: ok, rules follow inheritance semantics as methods do. how do I make a method in a deriving class say "don't care about the methods in the derived class. MMD stops here."?
[bjoern]: also lookahead, I think. 19:12
jnthn masak: I don't quite get what you want. Do you mean "don't look at the MMD in any parents?" 19:14
[bjoern] The idea being that if, say, <^ReservedName> was the complement of ReservedName you could write <Name> & <^ReservedName>, it would seem
masak jnthn: the point of MMD/inheritance is fallthrough to parent classes. but in the case of some methods I might want to suppress that. 19:15
jnthn masak: In which case, write a multi with a permissive enough signature to accept anything I guess, and do whatever is meant to happen in that case.
masak jnthn: ah, ok. guess the same trick works for a proto regex.
jnthn Perhaps yes, I don't really grok how close the relationship between multis and proto regexes is. 19:16
masak [bjoern]: I still think the solution you want is to be found in lookaheads.
[bjoern] Yeah but i find those a bit ugly and difficult to understand 19:17
jnthn I've kinda thought of it as a nice-ish way of thinking about it sometimes, but while regex = method is really a quite deep equivalence, I'm not sure protoregexen and multi dispatch would have that same depth of relationship.
masak jnthn: one difference between regexes-in-grammars and methods-in-classes is that the regexes can occur outside of grammars too. then they feel more like subs.
[bjoern] In some bnf variants you can just write, say, "Name - ReservedName"
jnthn masak: Well, annonymous methods maybe fill the same niche as outside-of-class regexes. 19:18
masak jnthn: yes, but regexes outside of grammars are named. and there are anonymous regexes, too. 19:20
jnthn Yes, true.
19:21 wknight8111 left 19:22 wknight8111 joined, dukeleto left
jnthn masak: Your 3d is pretty. :-) 19:23
woo, the YPAC::Asia submissions is open. 19:25
Now I just need to think of talks to submit.
19:26 ejs0 left
masak jnthn: thanks. working on making it interactive now. 19:26
jnthn cool 19:29
Matt-W Hurrah for pretty 19:34
masak yes; I will definitely try to focus on the pretty in my blogging from now on. 19:36
it seems to have a way of motivating people.
Su-Shee masak: isn't you blog within perlsphere?! 19:39
masak Su-Shee: no, I don't think so. 19:40
19:40 mvuets joined
Su-Shee masak: may I have the url, then, please? :) 19:40
19:41 ejs0 joined
masak Su-Shee: use.perl.org/~masak/journal/ use.perl.org/~masak/journal/rss 19:41
Su-Shee masak: funny, yesterday I googled around for some papers about MVC stuff.. 19:43
masak MVC is still a slightly nebulous concept to me. I think I get the basics, but... 19:44
Su-Shee masak: the wikipedia article got two papers linked I found helpful.
masak heads over to Wikipedia 19:45
Su-Shee masak: the one from Derek Greer and the "What is behind MVC?" 19:46
masak Su-Shee: thanks. I'll have a look. 19:47
Su-Shee masak: and this I forgot: oreilly.com/php/archive/mvc-intro.html 19:51
(ignore the php part there.. ;)
masak I always ignore the PHP part. :)
19:53 cartes joined 19:55 cartes left 20:00 mtve left 20:08 maettu left 20:09 M_o_C left 20:16 payload left, masak left, mberends left, rafl left, baest left, eternaleye left, christine left, antiphase left, charsbar_ left, PerlJam left, Tene left, PerlJam joined, payload joined, masak joined, mberends joined, rafl joined, baest joined, christine joined, eternaleye joined, charsbar_ joined, antiphase joined, Tene joined, irc.freenode.net sets mode: +oo PerlJam Tene 20:19 baest left, mberends left, antiphase left, payload left, rafl left, PerlJam left, masak left, christine left, eternaleye left, Tene left, charsbar_ left 20:20 masak joined 20:21 rafl joined, baest joined, eiro left 20:22 jlaire left, PerlJam joined, payload joined, mberends joined, christine joined, eternaleye joined, charsbar_ joined, antiphase joined, Tene joined, irc.freenode.net sets mode: +oo PerlJam Tene, jonathanturner left 20:27 eiro joined
bpa has anyone used Module::Pluggable::Fast from pugs? 20:34
20:34 mvuets left
bpa I'm trying to get instances of plugins from it, but I'm struggling with it 20:35
20:36 ejs joined
bpa I guess I'll try a different tack 20:43
if I have a class name in $class, how would I create a new instance from that? 20:44
eval doesn't seem to be working
jnthn orly? 20:47
rakudo: class Pivo { }; eval('Pivo').new.WHAT.say 20:48
p6eval rakudo 6c43f9: OUTPUT«sh: ./perl6: No such file or directory␤»
jnthn ?
rakudo: class Pivo { }; eval('Pivo').new.WHAT.say
p6eval rakudo 6c43f9: OUTPUT«sh: ./perl6: No such file or directory␤»
jnthn es ist kaput.
bpa: As I just did there, works for me locally though. 20:49
bpa I hadn't tried it that way
that's much cleaner than me trying eval("\$obj = Pivo.new")
payload s/kaput/kaputt/ 20:50
jnthn payload: Mein Deutsch ist so kaputt. ;-) 20:51
payload ^^
jnthn never actually learned it, just holidayed/visited Germany/Switzerland/Austria quite a bit and slurped some random bits up.
20:53 ejs0 left 20:54 Su-Shee left
arnsholt Whee! Criminal Minds opened with a Terry Pratchett quote this time =D 20:58
Oh, opps. Wrong window
21:00 desertm4x joined
masak really likes the end of perlmonks.org/?node_id=765916 21:00
thing is, when TimToady is making things up, they sometimes end up in the spec. :P 21:01
payload rakudo: 0,1...&[+] 21:07
p6eval rakudo 6c43f9: OUTPUT«Statement not terminated properly at line 2, near "...&[+]"␤in Main (src/gen_setting.pm:3225)␤»
masak payload: &[+] NYI, I think. 21:08
jnthn masak: aye, nor ... afaik.
Though it maybe happened while I was away.
masak jnthn: it did.
moritz_++
jnthn moritz_++ indeed...didn't that need laziness, or did we also get that now?
masak but, of course, this particular examples... aye.
speaking of the spec, the stowaway in the latest big spec update made me return to thinking about defined/undefined vs passed/not-passed in parameters, as well as initialized/not-initialized in attributes. I still think this question merits a solution. 21:09
jnthn Ouch. Over 2,000 photos from the trip to sort through... 21:10
masak the pains of the modern life. 21:11
jnthn masak: I'm interested in that stowaway too. Not really awake enough or back into Perl 6 mode enough to think about it yet.
21:11 M_o_C joined
masak no prob. 21:11
21:11 sjohnson left 21:12 sjohnson joined
masak I'm not going anywhere. :) 21:12
21:19 sjohnson left 21:23 sjohnson joined
eternaleye rakudo: .say for 0,1...&[+] 21:23
p6eval rakudo 6c43f9: OUTPUT«Statement not terminated properly at line 2, near "...&[+]"␤in Main (src/gen_setting.pm:3225)␤»
eternaleye rakudo: .say for 0,1...&infix:<+> 21:24
p6eval rakudo 6c43f9: OUTPUT«Null PMC access in find_method()␤in Main (/tmp/G94MMIkkx7:2)␤»
eternaleye masak: You know what to do!
masak eternaleye: we don't have lazy yet.
21:24 dukeleto_ is now known as dukeleto
masak eternaleye: variants of that one have been submitted already, I think. 21:24
eternaleye masak: Null PMC > *
masak: Okay
pmurias rakudo: say :(1,2,3) 21:25
p6eval rakudo 6c43f9: OUTPUT«Method 'named' not found for invocant of class 'Undef'␤»
masak pmurias: ooh! is that one new?
21:25 dukeleto left
eternaleye rakudo: say :(1,2,3).params 21:25
p6eval rakudo 6c43f9: OUTPUT«Null PMC access in find_method()␤in Main (/tmp/fdQBRz0WOb:2)␤»
pmurias rakudo: say (:($foo)).perl 21:26
jnthn rakudo: sub foo(1,2,3) { }
p6eval rakudo 6c43f9: OUTPUT«too few arguments passed (0) - 1 params expected␤in Main (/tmp/eBEjsYawep:0)␤»
rakudo 6c43f9: OUTPUT«Malformed routine definition at line 2, near "foo(1,2,3)"␤in Main (src/gen_setting.pm:3225)␤»
jnthn rakudo: say (:(1,2,3)).WHAT
p6eval rakudo 6c43f9: OUTPUT«Null PMC access in find_method()␤in Main (/tmp/ehmUcLv60H:2)␤»
masak ETOOMANYBUGS
pmurias pugs: sub foo($a,$b) {};say &foo.signature.params.perl 21:27
p6eval pugs: OUTPUT«*** No such subroutine: "&params"␤ at /tmp/yLJghBnYku line 1, column 23 - line 2, column 1␤»
jnthn Not entirely sure if we're attempting any :(...) signature syntax spec tests, mind.
pmurias jnthn: is &foo.signature.params speced?
jnthn So I suspect there's a bit of "standalone signatures NYI" here. 21:28
eternaleye pmurias: It worked for me yesterday
jnthn pmurias: I think there's general agreement signatures should be introspectable.
pmurias: Exactly how that looks I think is not yet fully spec'd.
pmurias jnthn: are the ones in rakudo introspectable?
jnthn pmurias: Aye.
eternaleye rakudo: sub foo( *@a ) {}; say &foo.singature.params.perl 21:29
p6eval rakudo 6c43f9: OUTPUT«Method 'singature' not found for invocant of class 'Perl6Sub'␤»
jnthn Though not through an itnerface that I think is gonna fly.
eternaleye rakudo: sub foo( *@a ) {}; say &foo.signature.params.perl
p6eval rakudo 6c43f9: OUTPUT«["named" => 0, "slurpy" => 1, "name" => "\@a", "multi_invocant" => 1, "nom_type" => Any, "type" => Any, "cons_type" => undef]␤»
eternaleye typos--
pmurias &foo.signature.params also work in mildew, but it's containts totally different stuff 21:30
* it contains
masak yawns
jnthn Maybe it's one of those implementation specific thingies.
eternaleye jnthn: I think what would make the most sense would probably be a List (indexed by position in the declaration, cf S06:767) of Hash (keyed by property, eg slurpy, positional, named, etc). This would allow an easy check for slurpies in, say, infix:<...> by using 'if +&generator.signature.params.grep( *.<slurpy> )'
jnthn eternaleye: list of hash is pretty much what Rakudo does.
eternaleye jnthn: Nope, Array of Pair
rakudo: sub foo( *@a ) {}; say &foo.signature.params.<slurpy> 21:31
p6eval rakudo 6c43f9: OUTPUT«named 0slurpy 1name @amulti_invocant 1nom_type Any()type Any()cons_type ␤»
eternaleye rakudo: sub foo( *@a ) {}; say &foo.signature.params.kv.perl
p6eval rakudo 6c43f9: OUTPUT«[0, "named" => 0, 1, "slurpy" => 1, 2, "name" => "\@a", 3, "multi_invocant" => 1, 4, "nom_type" => Any, 5, "type" => Any, 6, "cons_type" => undef]␤»
jnthn rakudo: sub foo($a, :$b) { }; &foo.signature.elems.say
p6eval rakudo 6c43f9: OUTPUT«1␤»
jnthn rakudo: sub foo($a, :$b) { }; &foo.signature.PARROT
p6eval rakudo 6c43f9: ( no output )
jnthn rakudo: sub foo($a, :$b) { }; say &foo.signature.PARROT
p6eval rakudo 6c43f9: OUTPUT«Signature␤»
jnthn rakudo: sub foo($a, :$b) { }; say &foo.signature.params.PARROT
p6eval rakudo 6c43f9: OUTPUT«ResizablePMCArray␤» 21:32
jnthn rakudo: sub foo($a, :$b) { }; say &foo.signature.params.elems
p6eval rakudo 6c43f9: OUTPUT«12␤»
jnthn oh, eww.
eternaleye Precisely.
jnthn Somewhere, it's getting flattened out.
rakudo: sub foo($a, :$b) { }; say &foo.signature.params[0].PARROT
p6eval rakudo 6c43f9: OUTPUT«Hash␤»
eternaleye Maybe a Capture of Hash, which would resist flattening? 21:33
jnthn Last I knew List and Capture were getting unified anyway.
No, I suspect $other_issue
eternaleye jnthn: True
jnthn I'm not quite sure what without looking more.
masak I'm off bug-submitting duty tonight.
jnthn I know that under the hood they are stored as an array of hash because that's how the multi-dispatcher introspects the signature for the stuff it wants.
masak anyone who finds a but can just submit it at will. :) 21:34
jnthn But clearly there's something wrong in how it's being exposed.
eternaleye jnthn++ # You'll find it. Anata wo shinjeru!
21:34 ejs left
masak what language is that? 21:34
jnthn I think Japanese.
eternaleye Japanese, but scim-- isn't working right now
jnthn What does shinjeru mean? 21:35
eternaleye The QT immodule crashes programs on input events
jnthn "Anata wo" is just "you" IIRC.
eternaleye "believe"
"I believe in you"
pmurias here's how signatures are implemented in mildew svn.pugscode.org/pugs/v6/re-mildew/...gnature.pm
eternaleye "anata" is 'you', "wo" marks the object in a sentence
jnthn wo marks the object? OK. 21:36
jnthn is trying to get the markers straight in his head.
Only started learning Japanese on a train yesterday. ;-)
eternaleye "wa" for the subject, "ga" for the topic (think given in p6)
jnthn Ah! OK.
masak seems a lot like Lojban to me. :) 21:37
jnthn I was wondering if "wo" was just the spelling for what I was hearning as "wa" but OK, they're different.
#perl6++ # Perl 6 *and* Japanese support.
eternaleye ni for non-relative time as well as for direction of motion (it shares the latter with "he" (pron. "eh"))
21:37 pmurias left
eternaleye "de" for static/current location 21:38
jnthn Are things like "ka" (question marker) also in that set of thingies?
eternaleye masak: I want to learn Lojban, but I haven't had the time and nobody I know locally (IE near enough to speak aloud) knows it
jnthn: Yes. That set is called "particle" words 21:39
masak eternaleye: I suspect one day I'll learn Lojban, too.
eternaleye masak: Go for the conlang trifecta, learn Klingon too!
masak meh. :) 21:40
being able to speak Ido and Interlingua fluently would be nice, though.
and Lingua Franca Nova looks cute.
jnthn Are conlangs generally easier to learn than natural languages? 21:41
masak if they're designed right, yes.
and auxlangs are often designed with ease in mind. 21:42
not all conlangs, though.
certainly not Klingon. :)
eternaleye jnthn: They're much more consistent - you can learn a ruleset rather than a ruleset + exceptions
(in general)
masak I've been spending the day speaking Esperanto. I'm fortunate enough to be living close to a native speaker. 21:43
21:44 mvuets joined
eternaleye Languages are _fun_. I think that's why I had trouble with Learning Python for my first programming language, but Perl stuck really easily as my firs proglang 21:44
Perl is a much more _human_ language.
masak I had a uni lecture about Perl 5 two days ago. the title was 'Perl -- the programming language for humans'. 21:45
21:45 desertm4x left
mvuets perl6: my @a = <1 3 5>; say "a is @a[]"; 21:49
p6eval pugs: OUTPUT«a is 1 3 5␤»
..elf 27287, rakudo 6c43f9: OUTPUT«a is @a[]␤»
jnthn mvuets: hi! :-)
mvuets jnthn, hey Jonathan! (: How are you?
jnthn mvuets: pugs is more correct in this case I think. Rakudo doesn't implement those postfixes yet.
eternaleye rakudo: my @a = <1 3 5>; say "a is { @a[] }";
p6eval rakudo 6c43f9: OUTPUT«a is 1 3 5␤»
jnthn mvuets: I'm good, thanks...you? 21:50
masak it's blocking on some string parsing, right?
mvuets jnthn, excelent, thanks. As you see, wanna learn Perl 6 (: Hope this friendly community will help (;
eternaleye, oh, thanks. 21:51
eternaleye mvuets: No problem. Rakudo does fine interpolating closures, so it's a useful workaround.
mvuets jnthn, i have only Rakudo now. Thought something wrong with it. Now see that it's just not implemented yet. 21:52
jnthn mvuets: Oh, I'm sure it will. :-)
Just a Not Yet Implemented thingy.
masak: Not 100% sure what it blocks on, but I'd guess that. 21:53
I kinda leave parsing stuff to pm, he's far better at it than me.
masak he's far better at it than most of us.
bpa A quick learning perl6 question: if you return %hash, does that pass by value or ref? 21:54
masak he implemented a Perl 6 grammar lib in 16k of assembly! :)
16kloc, I mean.
bpa: good question. I've never been in a position to need to distinguish.
I'd guess by value, though.
mvuets {} within "" is just interpolated kinda in-line perl or are there some limitations? 21:55
masak rakudo: my %h = 'foo' => 'bar'; sub foo(%h is rw) { return %h }; my %h2 = foo(%h); %h<foo> = 'baz'; say %h.perl 21:56
p6eval rakudo 6c43f9: OUTPUT«{"foo" => "baz"}␤»
masak bpa: seems it passes it by ref.
eternaleye rakudo: my @seed = 0, 1; my @fibs = @seed ... { state $count //= 0; $count++ > 100 ?? () !! $^a + $^b ; }; .say for @fibs 21:58
p6eval rakudo 6c43f9: OUTPUT«Null PMC access in find_method()␤in sub infix:... (/tmp/3LXeEOFQs9:2)␤called from Main (/tmp/3LXeEOFQs9:2)␤»
eternaleye 0.o 21:59
21:59 wknight8111 left
masak I think moritz_++ mentioned that bug. 22:00
eternaleye I could have sworn that worked yesterday, almost exactly the same (although it was (state $something)++, which I changed to avoid a "use of undef" warning) 22:01
Can't remember who did it though 22:02
masak I wish there was a way to go back to previous IRC conversations...
'night, all. 22:03
22:03 masak left
bpa pugs seems to pass by value 22:04
eternaleye rakudo: say 1,2...{ (state $i)++ < 100 ?? $^a + $^b !! () } 22:06
p6eval rakudo 6c43f9: OUTPUT«Use of uninitialized
..value␤123581321345589144233377610987159725844181676510946177112865746368750251213931964183178115142298320401346269217830935245785702887922746514930352241578173908816963245986102334155165580141267914296433494437701408733113490317018363119032971215073480752697677…
eternaleye (That was what TimToady entered) 22:07
rakudo: my @seed = 0, 1; my @fibs = @seed ... { state $count //= 0; $count++ < 100 ?? $^a + $^b !! () }; .say for @fibs
p6eval rakudo 6c43f9: OUTPUT«Null PMC access in find_method()␤in sub infix:... (/tmp/ZlgW0ndK9v:2)␤called from Main (/tmp/ZlgW0ndK9v:2)␤»
eternaleye rakudo: my @seed = 0, 1; my @fibs = @seed ... { state $count //= 0; my @foo = $count++ < 100 ?? $^a + $^b !! (); say ~@foo; return @foo }; 22:08
p6eval rakudo 6c43f9: OUTPUT«Null PMC access in find_method()␤in sub infix:... (/tmp/6uvtNBCbMo:2)␤called from Main (/tmp/6uvtNBCbMo:2)␤»
eternaleye rakudo: my @seed = 0, 1; my @fibs = @seed ... { (state $count)++ > 100 ?? () !! $^a + $^b ; }; .say for @fibs 22:09
p6eval rakudo 6c43f9: OUTPUT«Use of uninitialized
..value␤0␤1␤1␤2␤3␤5␤8␤13␤21␤34␤55␤89␤144␤233␤377␤610␤987␤1597␤2584␤4181␤6765␤10946␤17711␤28657␤46368␤75025␤121393␤196418␤317811␤514229␤832040␤1346269␤2178309␤3524578␤5702887␤9227465␤14930352␤24157817␤39088169␤63245986␤102334155␤165580141␤267914296␤433494437␤70140…
eternaleye Hm, so it's //= that does it
mvuets pugs: my @a = <1 3 5>; say "a is @a"; 22:10
p6eval pugs: OUTPUT«a is @a␤»
eternaleye rakudo: my @seed = 0, 1; my @fibs = @seed ... { state $count = $count // 0; $count++ > 100 ?? () !! $^a + $^b ; }; .say for @fibs
p6eval rakudo 6c43f9:
..OUTPUT«0␤1␤1␤2␤3␤5␤8␤13␤21␤34␤55␤89␤144␤233␤377␤610␤987␤1597␤2584␤4181␤6765␤10946␤17711␤28657␤46368␤75025␤121393␤196418␤317811␤514229␤832040␤1346269␤2178309␤3524578␤5702887␤9227465␤14930352␤24157817␤39088169␤63245986␤102334155␤165580141␤267914296␤433494437␤701408733␤1134903170␤1836311903␤…
22:11 DemoFreak left
mvuets Do you know any russian blogs which describe Perl 6 alike it is done by moritz_'es perlgeek.de blog? perl6.ru is out of date. 22:14
obra_ Where are the ENV variables Rakudo pays attention to documented? 22:15
22:15 pmurias joined
obra_ (Specifically, I'm looking for how to set @*INC from the commandline) 22:15
eternaleye phenny: ask moritz_ Is it known that &infix:<...> fails on //=, but = $a // $b works fine?
phenny eternaleye: I'll pass that on when moritz_ is around.
obra_ my assumption is that it's PERL6LIB 22:16
eternaleye phenny: tell moritz_ example: my @seed = 0, 1; my @fibs = @seed ... { state $count //= 0; $count++ < 100 ?? $^a + $^b !! () }; .say for @fibs
phenny eternaleye: I'll pass that on when moritz_ is around.
22:22 sri_kraih_ joined 22:25 buubot left 22:30 buubot joined, buubot left
mvuets I see there's such cool thing as non-ascii alias to hyper-operator, i mean e.g. »+«. Why not to add → (\u2192) as an alias to "->"? I saw Scala uses one. 22:31
22:31 vv31 left 22:34 iblechbot left, buubot joined, buubot left 22:36 buubot joined 22:39 sri_kraih left 22:42 payload left, Eevee left 22:43 fridim left 22:44 pmurias left 22:45 dduncan joined 22:47 Eevee joined 22:49 PacoLinux left
lisppaste3 mvuets pasted "new lambda operator looks pretty cool (:" at paste.lisp.org/display/82632 22:49
22:49 PacoLinux joined 22:52 buubot left, buubot joined 22:54 snarkyboojum joined 22:58 jlmoko joined 23:07 payload joined 23:17 bpa left 23:19 M_o_C left
mvuets perl6: say 'match' if 'aéo' ~~ m:a/aeo/; 23:22
p6eval elf 27287: ( no output )
..pugs: OUTPUT«Error eval perl5: "if (!$INC{'Pugs/Runtime/Match/HsBridge.pm'}) {␤ unshift @INC, '/home/evalenv/pugs/perl5/Pugs-Compiler-Rule/lib';␤ unshift @INC, '/home/evalenv/pugs/third-party/Parse-Yapp/lib';␤ eval q[require 'Pugs/Runtime/Match/HsBridge.pm'] or die $@;␤}␤'Pugs::Runtime…
..rakudo 6c43f9: OUTPUT«Colons cannot be used as delimiters in quoting constructs at line 2, near ":a/aeo/;"␤in Main (src/gen_setting.pm:3225)␤»
23:24 viklund left
mvuets perl6: say 'match' if 'aEo' ~~ m:i/aeo/; 23:24
p6eval rakudo 6c43f9: OUTPUT«Colons cannot be used as delimiters in quoting constructs at line 2, near ":i/aeo/;"␤in Main (src/gen_setting.pm:3225)␤»
..elf 27287: ( no output )
..pugs: OUTPUT«Error eval perl5: "if (!$INC{'Pugs/Runtime/Match/HsBridge.pm'}) {␤ unshift @INC, '/home/evalenv/pugs/perl5/Pugs-Compiler-Rule/lib';␤ unshift @INC, '/home/evalenv/pugs/third-party/Parse-Yapp/lib';␤ eval q[require 'Pugs/Runtime/Match/HsBridge.pm'] or die $@;␤}␤'Pugs::Runtime…
mvuets perl6: say 'match' if 'aeo' ~~ m/aeo/; 23:26
p6eval elf 27287, rakudo 6c43f9: OUTPUT«match␤»
..pugs: OUTPUT«Error eval perl5: "if (!$INC{'Pugs/Runtime/Match/HsBridge.pm'}) {␤ unshift @INC, '/home/evalenv/pugs/perl5/Pugs-Compiler-Rule/lib';␤ unshift @INC, '/home/evalenv/pugs/third-party/Parse-Yapp/lib';␤ eval q[require 'Pugs/Runtime/Match/HsBridge.pm'] or die $@;␤}␤'Pugs::Runtime…
23:30 mvuets left 23:34 mikehh joined, simcop2387 left 23:54 Muixirt left, last left 23:55 justatheory joined 23:57 nErVe joined